JP5365440B2 - Image processing apparatus and image processing program - Google Patents

Abstract

An image processing apparatus includes an image receiving section, a succession value image generating section and a solid converting section. The image receiving section receives an image contains a line having breaks. The succession value image generating section generates a succession value image having as a pixel value the number of times of succession of black or white pixels in the image received by the image receiving section. The solid-line converting section, based on the succession value image generated by the succession value image generating section, performs solid-line conversion in which the line having breaks is changed into a solid line.

Description

  The present invention relates to an image processing apparatus and an image processing program.

Image processing is performed on a document including a table or the like formed by a line having a break such as a dotted line.
As a technique related to this, for example, Patent Document 1 discloses that a document image is extracted from a document image for the purpose of extracting a discontinuous ruled line such as a broken line or a chain line of a table in a document image with a simple process. Scans in the form of a binary original image, scans the read binary original image in each of the horizontal and vertical directions, detects a black bit string and a white bit string connected thereto, and detects the detected black bit string and the connected black bit string The length of the white bit string is determined with respect to the white bit string, the black bit string whose length is less than or equal to the threshold and the white bit string less than the threshold connected thereto are detected, and the detected black bit string and the detected black bit string are connected to the detected black bit string. A basic bit string is detected for a series of white bit strings that are less than or equal to a threshold value, a limited repeated bit string in which the detected basic bit string is regularly repeated a predetermined number of times or more is detected, and a limited repeated pattern is extracted. There has been disclosed.

  Further, for example, Patent Document 2 provides a table recognition device that can accurately recognize table structures and characters in cells from graphic data read from printed matter / documents, etc., regardless of characters, broken lines, dotted lines, and line types. A black pixel labeling unit that creates a circumscribed rectangle list of connected black pixels of graphic data, and a character rectangle estimation unit that determines whether the inside of the rectangle is a character from the length of a side of the circumscribed rectangle A reduced image creation unit for creating a reduced image of a portion excluding black pixels in a rectangle estimated to be a character, and a black pixel run in two vertical and horizontal directions of the reduced image having a certain length or more A run extraction unit for extracting a ruled line, a ruled line extraction unit for extracting a ruled line by concatenating the extracted runs, a table structure extraction unit for searching a rectangle surrounded by the extracted ruled line and extracting it as a table cell It is disclosed that it has .

  Further, for example, in Patent Document 3, a binary image of a document is input from a binary image input unit to a binary image memory for the purpose of recognizing a dotted ruled line and processing a table including the dotted ruled line. The rectangle extraction unit extracts a rectangle including black pixels connected in relation to the ruled line direction, the dotted line element extraction unit extracts a rectangle smaller than a predetermined size as a dotted line element, and the dotted line element extraction unit relates to the ruled line direction in advance. It is disclosed that dotted line elements that are closer than a predetermined interval are integrated and extracted as a dotted rectangle.

  In addition, for example, Patent Document 4 aims to provide a ruled line recognition method for enabling accurate processing even for a table including a non-solid ruled line such as a dotted line. The black run and the white run of each line in the main scanning direction and the sub-scanning direction are extracted, the histogram for the length of each of the extracted black run and the white run is obtained, and the value of the peak portion of the histogram is constant. When the value is exceeded, a continuous range of black runs and white runs having a length corresponding to the peak portion is extracted as a candidate for a non-solid ruled line, and the threshold processing for the length of the extracted black run is performed. A ruled line recognition characterized in that the black run of the above length is extracted as a candidate for a solid line, and the extracted non-solid line and solid line rule candidates are combined together and recognized as a ruled line. Way It is shown.

Japanese Patent Laid-Open No. 05-040847 JP 05-012489 A Japanese Patent Application Laid-Open No. 07-230525 Japanese Patent No. 3140079

  The present invention relates to an image processing apparatus and image processing in which when a line having a break in an image is turned into a solid line, a line other than the line having a break is prevented from being turned into a solid line as a line having a break. The purpose is to provide a program.

The gist of the present invention for achieving the object lies in the inventions of the following items.
According to the first aspect of the present invention, there is provided an image receiving means for receiving an image, and a continuous frequency image creation for creating a continuous power image having a frequency of black pixels or white pixels in the image received by the image receiving means as a pixel value. And a solid line forming means for performing a solid line changing a broken line to a solid line based on the continuous frequency image created by the continuous frequency image creating means, and the solid line drawing means includes the continuous frequency image. By comparing a pixel value of a pixel having a frequency that is continuous in the horizontal direction or the vertical direction as a pixel value with a predetermined first threshold value. The line segment extracted and sandwiched between the two first pixels intersects the direction of the power at a right angle, and is determined in advance as the pixel value of the pixel sandwiched between the two first pixels. Second threshold To extract a second pixel located on a line segment sandwiched between the first pixels, and use the second pixel as a broken line portion of the broken line to make a solid line. an image processing apparatus, characterized in that there.

According to a second aspect of the present invention, a pixel that is in contact with a solid line or a broken line in an image received by the receiving unit based on an image created based on the solid line formed by the solid line forming unit. The image processing apparatus according to claim 1, further comprising separation means for separating a lump.

According to a third aspect of the present invention, there is provided a continuous white power image creating means for creating a continuous white power image having a frequency of continuous white pixels in an image subjected to a solid line by the solid line forming means as a pixel value; A continuous black power image creating means for creating a black continuous power image having a frequency of continuous black pixels in the image that has been converted into a solid line by the means, and a white continuous image created by the white continuous power image creating means Further comprising line segment extraction means for extracting a line segment from the image received by the image receiving means based on the frequency image and the black continuous frequency image created by the black continuous frequency image creating means. An image processing apparatus according to claim 1.

According to the invention of claim 4 , the computer generates an image reception means for receiving an image and a continuous frequency image having as a pixel value the frequency at which black pixels or white pixels in the image received by the image reception means are continuous. Based on the frequency image creating means and the continuous frequency image created by the continuous frequency image creating means, the frequency line creating means functions as a solid line forming means for changing to a solid line, and the solid line forming means includes By comparing the pixel value of a pixel in the continuous power image having a pixel value that is a frequency that is continuous in the horizontal direction or the vertical direction, the two first values are compared with each other. The line segment sandwiched between the two first pixels intersects the direction of the power at a right angle, and the pixel value of the pixel sandwiched between the two first pixels A second pixel located on a line segment sandwiched between the first pixels is extracted by comparing the second threshold value with a second threshold value determined for the second pixel, and the second pixel is defined as a break in the line having the break. An image processing program characterized by performing solid line as a part .

According to the image processing apparatus of the first aspect, when a line with a break in the image is made a solid line, it can be suppressed that a line other than the line with a break is made into a solid line as a line with a break. And compared with the case where it does not have this structure, the load of the process which makes a continuous line a solid line can be reduced.

According to the image processing apparatus of the second aspect , it is possible to separate the pixel block in contact with the line.

According to the image processing apparatus of the third aspect , when a line segment is extracted from the image, it is possible to suppress extraction of a non-line segment as a line segment.

According to the image processing program of the fourth aspect , when a line with a break in the image is made into a solid line, it can be suppressed that a line other than the line with a break is made into a solid line as a line with a break. And compared with the case where it does not have this structure, the load of the process which makes a continuous line a solid line can be reduced.

It is a conceptual module block diagram about the structural example of 1st Embodiment. It is a detailed module block diagram about the structural example of 1st Embodiment. It is explanatory drawing which shows the example of a reception image. It is explanatory drawing which shows the example of a white run map image. It is explanatory drawing which shows the example of the property of a white run map image. It is explanatory drawing which shows the example of the property of a white run map image. It is explanatory drawing which shows the process example by the white run map filtering module. It is explanatory drawing which shows the example of a process by the logical sum process module V etc. It is explanatory drawing which shows the example of a process by the black run map image creation module H etc. It is explanatory drawing which shows the process example by a logical sum processing module. It is explanatory drawing which shows the process example by an inversion processing module. It is explanatory drawing which shows the process example by 1st Embodiment. It is a conceptual module block diagram about the structural example of 2nd Embodiment. It is a detailed module block diagram about the structural example of 2nd Embodiment. It is explanatory drawing which shows the example of a relationship between black run length and white run length. It is explanatory drawing which shows the process example by the black-and-white run map filtering module V. FIG. It is explanatory drawing which shows the process example by 2nd Embodiment. It is a detailed module block diagram about the structural example of the inversion and logical product processing module of 3rd Embodiment. It is a block diagram which shows the hardware structural example of the computer which implement | achieves 1st-3rd embodiment.

Hereinafter, examples of various preferred embodiments for realizing the present invention will be described with reference to the drawings.
FIG. 1 is a conceptual module configuration diagram of a configuration example according to the first embodiment.
The module generally refers to components such as software (computer program) and hardware that can be logically separated. Therefore, the module in the present embodiment indicates not only a module in a computer program but also a module in a hardware configuration. Therefore, the present embodiment also serves as an explanation of a computer program, a system, and a method. However, for the sake of explanation, the words “store”, “store”, and equivalents thereof are used. However, when the embodiment is a computer program, these words are stored in a storage device or stored in memory. It is the control to be stored in the device. Modules may correspond to functions one-to-one, but in mounting, one module may be configured by one program, or a plurality of modules may be configured by one program, and conversely, one module May be composed of a plurality of programs. The plurality of modules may be executed by one computer, or one module may be executed by a plurality of computers in a distributed or parallel environment. Note that one module may include other modules. Hereinafter, “connection” is used not only for physical connection but also for logical connection (data exchange, instruction, reference relationship between data, etc.).
In addition, the system or device is configured by connecting a plurality of computers, hardware, devices, and the like by communication means such as a network (including one-to-one correspondence communication connection), etc., and one computer, hardware, device. The case where it implement | achieves by etc. is included. “Apparatus” and “system” are used as synonymous terms. Of course, the “system” does not include a social “mechanism” (social system) that is an artificial arrangement. “Predetermined” means that the process is determined before the target process, and not only before the process according to this embodiment starts but also after the process according to this embodiment starts. In addition, if it is before the target processing, it is used in accordance with the situation / state at that time or with the intention to be decided according to the situation / state up to that point.

  As illustrated in FIG. 1, the image processing apparatus according to the first embodiment includes an image reception module 110, a table ruled line correction module 120, a contact character separation module 130, a logical sum synthesis module 140, and an image output module 150. doing.

  The image reception module 110 is connected to the front ruled line correction module 120 and the contact character separation module 130. The image is received, and the image is transferred to the table ruled line correction module 120 and the contact character separation module 130. Accepting an image means, for example, reading an image with a scanner, a camera, etc., receiving an image from an external device via a communication line by fax, etc., a hard disk (in addition to what is built in a computer, a communication line For example, reading out an image stored in a device etc.). Here, the received image is a binary image, but may be a multi-valued image (including a color image). In the case of a multi-value image, binarization processing is performed. One image may be received or a plurality of images may be received. Further, as long as the content of the image includes a ruled line constituting a table or the like, it may be a document used for business, a form, a pamphlet for advertisement, or the like. Further, the ruled line may include a line with a break, or the pixel block may be in contact with the ruled line.

Here, the broken line is a line other than a solid line, and includes, for example, a dotted line, a broken line, a one-dot chain line, a two-dot chain line, and the like, and a line in which these are mixed. Hereinafter, it is also collectively referred to as a dotted line.
The pixel block includes at least a pixel area other than the ruled line and continuous in four or eight connections, and includes a set of these pixel areas. The set of these pixel areas means that there are a plurality of continuous pixel areas such as 4-connected, and the plurality of pixel areas are in the vicinity. Here, what is in the vicinity is, for example, an image area in which the pixel areas are close to each other in distance, an image area that is projected vertically or horizontally so as to cut out one character at a time from a line as a sentence, and cut out at a blank spot Or an image region cut out at a predetermined interval.
In many cases, an image of one character is formed as one pixel block. However, it is not necessary that the pixel area is actually recognizable as a character by humans. There are a part of a character, a pixel region that does not form a character, and the like, and any pixel block may be used.
Further, the ruled line is generally black, but can also correspond to a white ruled line. However, in the following description, the case of a black ruled line will be mainly described.

The table ruled line correction module 120 is connected to the image reception module 110, the contact character separation module 130, and the logical sum synthesis module 140. The table ruled line correction module 120 determines the frequency of black pixels or white pixels in the image received by the image reception module 110. A continuous frequency image having pixel values is created. Next, based on the created continuous frequency image, a solid line is formed by changing a line having a break into a solid line. Then, an image obtained by making the line with the cut line a solid line is passed to the contact character separation module 130 and the logical sum synthesis module 140. Further, the image to be transferred to the contact character separation module 130 or the like may be an image having only a solid ruled line. That is, an image obtained by extracting only the ruled line portion from the image received by the image receiving module 110 (an image obtained by removing the pixel block) may be used.
For example, the table ruled line correction module 120 has two first pixels extracted in comparison with a predetermined first threshold that are pixels in the continuous frequency image as both ends, and the first pixel When a pixel in between is a second pixel extracted by comparison with a predetermined second threshold value, the second pixel is used as a solid line as a broken line. May be. Details of the configuration and processing contents in the table ruled line correction module 120 will be described later with reference to FIG.

The continuous power image is an image having, as a pixel value, a frequency (a so-called run length) in which pixels of the same color as the target pixel are continuous in the horizontal direction or the vertical direction. In other words, the value of the run length is given as a pixel value to the pixels constituting the run length. Therefore, the pixels in the run length have the same pixel value. Hereinafter, the continuous power image is also referred to as a run map image. An image based on the run length of black pixels is called a black run map image, and an image based on the run length of white pixels is also called a white run map image.
This run map image has the same size (width × vertical length) as the image received by the image receiving module 110. However, the image received by the image receiving module 110 is a binary image, but the run map image is a multi-valued image in which one pixel can represent the frequency. For example, it may be 8 bits / pixel. In this case, when the frequency is 256 or more, the pixel value may be fixed to 255.
The solid line means that a line having a break is changed to a solid line. Specifically, the line having a break is connected (filled with a space between short lines) to become a solid line.

  The contact character separation module 130 is connected to the image reception module 110, the table ruled line correction module 120, and the logical sum synthesis module 140. Based on the image created based on the solid line formed by the table ruled line correction module 120, the pixel block that is in contact with the solid line or the broken line in the image received by the image receiving module 110 is separated. Specifically, for example, an image having only ruled lines is subjected to inversion processing (processing for changing white pixels to black pixels and black pixels to white pixels), and contraction processing is performed on the inverted image. A logical product process with the image received by the image receiving module 110 is performed on the image that has been subjected to the inversion process and the contraction process, and the pixel block that is in contact is separated from the ruled line that is a solid line or a broken line To do. Then, an image of only the image other than the ruled line (image of only the character) is passed to the logical sum synthesis module 140. Details of the configuration, processing contents, and the like in the contact character separation module 130 will be described later with reference to FIG.

  The logical sum synthesis module 140 is connected to the table ruled line correction module 120, the contact character separation module 130, and the image output module 150. An image of only the ruled line portion is received from the table ruled line correction module 120, and only the image of the ruled line portion is received, and the pixel block contacting the ruled line is separated from the contact character separating module 130, and only the image other than the ruled line is received. These images are received, and these are combined by logical sum to create a solid line from the cut line and separate the pixel block in contact with the ruled line. Then, the image is passed to the image output module 150. Details of the configuration, processing contents, etc. in the logical sum synthesis module 140 will be described later with reference to FIG.

  The image output module 150 is connected to the logical sum synthesis module 140. An image logically synthesized by the logical sum synthesis module 140 is received and the image is output. To output an image is, for example, printing on a printing device such as a printer, displaying on a display device such as a display, transmitting an image on an image transmission device such as a fax, or image to an image storage device such as an image database. , Storing in a storage medium such as a memory card, passing to another information processing apparatus, and the like. For example, the image may be output to a character recognition processing device, and the character recognition processing device may recognize the image and convert the image into a document such as XML (extensible Markup Language).

FIG. 2 is a detailed module configuration diagram for the configuration example of the first exemplary embodiment.
The image reception module 110 is connected to the white run map image creation module 222H, the logical sum processing module 224H, the white run map image creation module 222V, the logical sum processing module 224V, and the logical product processing module 233. As an image to be received, there is a received image 300 illustrated in FIG. Like an area 310 in the received image 300, there are contact occurrence areas 311 and 312 where characters and ruled lines are in contact (see FIG. 3B). In addition to the solid line 321 as a ruled line, there are a dotted line 322, a broken line 323, an alternate long and short dash line 324, etc. as in the region 320 in the received image 300 (see FIG. 3C).

  The table ruled line correction module 120 includes a horizontal direction processing module 221H, a vertical direction processing module 221V, and a logical sum processing module 227. The horizontal processing module 221H includes a white run map image creation module 222H, a white run map filtering module 223H, a logical sum processing module 224H, a black run map image creation module 225H, and a threshold processing module 226H. The vertical processing module 221V has a white run map image creation module 222V, a white run map filtering module 223V, a logical sum processing module 224V, a black run map image creation module 225V, and a threshold processing module 226V. The horizontal processing module 221H performs horizontal processing (processing for extracting horizontal ruled lines) on the image received by the image receiving module 110, and the vertical processing module 221V performs processing on the image received by the image receiving module 110. The processing in the vertical direction (processing for extracting vertical ruled lines) is different, and both are different in that they handle ruled lines in the horizontal and vertical directions, but the same processing is performed if rotated 90 degrees. , Have an equivalent corresponding module. In the following description, either one or both will be described.

The white run map image creation module 222 is connected to the image reception module 110 and the white run map filtering module 223. A white run map image is created using the image received from the image receiving module 110. The generated white run map image is passed to the white run map filtering module 223.
For example, specifically, the white pixel of the image is specified as the target pixel, and the run length of each white pixel (in the case of the processing of the white run map image creation module 222H, the number of white pixels connected in the horizontal direction, the white run In the case of the processing of the map image creation module 222V, the number of white pixels connected in the vertical direction) is counted, and a value having the run length length as the pixel value is given to the white pixels belonging to the run length. More specifically, when the run length of a block of white pixels is 10, the pixel value of 10 white pixels belonging to the block is set to 10.
FIG. 4 is an explanatory diagram illustrating an example of a white run map image. The white run map image 410 illustrated in FIG. 4A is created by the white run map image creation module 222H, and the white run map image 420 illustrated in FIG. 4B is created as a white run map image. This is created by the module 222V. In this figure, the pixel whose white run length length value is larger as it is blacker. In other words, the longer white run length is displayed in black. Note that the run length exceeding 255 is set to 255.

The white run map filtering module 223 is connected to the white run map image creation module 222 and the logical sum processing module 224. A filtering process is performed on the white run map image received from the white run map image creation module 222. Then, the result of the filtering process is passed to the logical sum processing module 224.
Here, the property of the white run map image will be described with reference to the examples of FIGS.
The example of the white run map image in FIG. 5B is for the character image exemplified in FIG. 5A by the white run map image creation module 222V, and the white run map image in FIG. 6B. An example of the image is an image of a dotted line illustrated in FIG. 6A by the white run map image creation module 222V. Thus, there are many relatively short white run pixels in the character pixel adjacent region, but there are many relatively long white run length pixels in the dotted line adjacent region. The white run map filtering module 223 uses this property.

FIG. 7 is an explanatory diagram showing an example of processing by the white run map filtering module 223V. The image illustrated in FIG. 7A is a dotted image that changes in black and white every two pixels with a width of two pixels. This image has the following properties.
(1) A long white run length exists beside the dotted line. In the example of FIG. 7A, the pixels at both ends (the rectangles indicated by the upward right lines, for example, the pixel 701) have a long white run length.
(2) The white pixels constituting the dotted line are short. In the example of FIG. 7A, a white pixel (a rectangle indicated by a right-down line, for example, the pixel 702) sandwiched between black pixels (pixel 703) constituting a dotted line has a short white run length. is there.

The white run map filtering module 223V detects white pixels constituting a dotted line using a horizontal 9 × 1 pixel filter (filter 710 illustrated in FIG. 7B). That is, for the white run map image, a pixel B having a short white run length sandwiched between pixels A having a long white run length is output as a black pixel (ON). In other cases, the pixel is output as a white pixel (OFF). The pixel A having a long white run length is a pixel having a pixel value larger than a predetermined threshold (Th_w), and the pixel B having a short white run length is a pixel having a short white run length. A pixel whose pixel value is smaller than a predetermined threshold (Th_s).
By this filtering process, a binary image is created in which only white pixels constituting the dotted line are turned on and other pixels are turned off (character and solid line).
The white run map filtering module 223H also performs the same filter processing using a vertical 1 × 9 pixel filter.
Note that the size of the filter may be variable so as to correspond to the detected dotted line width. The 9 × 1 pixel filter described above can detect a dotted line up to 7 pixels.

The logical sum processing module 224 is connected to the image receiving module 110, the white run map filtering module 223, and the black run map image creating module 225. An logical sum process is performed on the image received from the image reception module 110 and the white run map image received from the white run map filtering module 223 to create an image in which the dotted line is a solid line. Then, the created image is transferred to the black run map image creation module 225.
FIG. 8 is an explanatory diagram showing an example of processing by the logical sum processing module 224V or the like. The received image 300 is an image received by the image receiving module 110, and there are a dotted line, a broken line, and an alternate long and short dash line in the areas 811 812 812, respectively. The white run map image 420 is a white run map image created by the white run map image creation module 222V. For this white run map image 420, the white run map filtering module 223V creates a binary image in which white pixels constituting the dotted line are turned on and other pixels are turned off. A solid line image (vertical direction) 800 is an image created by the logical sum processing module 224V, and dotted lines and the like in the regions 821, 822, and 823 corresponding to the regions 811, 812, and 813 are solid lines.

  The black run map image creation module 225 is connected to the logical sum processing module 224 and the threshold processing module 226. A black run map image is created using the image received from the logical sum processing module 224 (image obtained by converting the dotted line into a solid line). Then, the created black run map image is transferred to the threshold processing module 226. The creation of the black run map image is equivalent to the creation of the white run map image of the white run map image creation module 222, and the black pixel of the image is specified as the target pixel, and the run length of each black pixel (black run map image) In the case of the processing of the creation module 225H, the number of black pixels connected in the horizontal direction is counted, and in the case of the processing of the black run map image creation module 225V, the number of black pixels connected in the vertical direction is counted. A value having a run length length as a pixel value is given to the pixel.

The threshold processing module 226 is connected to the black run map image creation module 225 and the logical sum processing module 227. Threshold processing is performed on each pixel in the black run map image received from the black run map image creation module 225. Then, the image subjected to the threshold processing is transferred to the logical sum processing module 227. Here, the threshold processing is to compare a predetermined threshold value with the pixel value of each pixel and extract pixels that are equal to or higher than the threshold value as pixels constituting a ruled line. . Therefore, the ruled lines exceeding the threshold value remain, and the ruled lines less than the threshold value are removed.
FIG. 9 is an explanatory diagram showing an example of processing by the black run map image creation module 225 and the threshold processing module 226.
A solid line image (horizontal direction) 900 illustrated in FIG. 9A is an output image of the logical sum processing module 224H. A black run map image creation module 225H, The result of processing by the threshold processing module 226H is a horizontal ruled line image 910. In this way, only horizontal ruled lines that are equal to or greater than the threshold value are extracted from an image obtained by converting horizontal dotted lines into solid lines.
The same processing by the black run map image creation module 225V and the threshold processing module 226V is also performed on the solid line image (vertical direction) 800 illustrated in FIG. 9B, and the processing result is the vertical ruled line image 920. .

The logical sum processing module 227 is connected to the threshold processing module 226H, the threshold processing module 226V, the inversion processing module 231, and the logical sum processing module 241. An image consisting only of horizontal ruled lines is received from the threshold processing module 226H, an image consisting only of vertical ruled lines is received from the threshold processing module 226V, and the two images are ORed. Then, the logically synthesized image is transferred to the inversion processing module 231 and the logical sum processing module 241.
FIG. 10 is an explanatory diagram showing an example of processing by the logical sum processing module 227. The logical sum processing module 227 receives the horizontal ruled line image 910 from the threshold processing module 226H and the vertical ruled line image 920 from the threshold processing module 226V, and creates a ruled line image 1000 by performing a logical sum synthesis.

The contact character separation module 130 includes an inversion processing module 231, a contraction processing module 232, and a logical product processing module 233.
The inversion processing module 231 is connected to the logical sum processing module 227 and the contraction processing module 232. An image having only ruled lines is received from the logical sum processing module 227, and reverse processing (logical negation processing) is performed. Then, the inverted image is transferred to the contraction processing module 232.
The contraction processing module 232 is connected to the inversion processing module 231 and the logical product processing module 233. The reversal-processed image is received from the reversal processing module 231 and contraction processing is performed. Then, the contracted image is transferred to the logical product processing module 233. Here, the contraction process is a process of converting the value of the pixel at the boundary of the black area into the value of the white pixel of the background component and reducing the black area by one or more pixels. For example, logical product processing of an image moved by one pixel and the original image may be performed. In addition, a contraction process as a conventionally known image process may be performed. As a result, an image for separating the character in contact with the ruled line is created.
The logical product processing module 233 is connected to the image reception module 110, the contraction processing module 232, and the logical sum processing module 241. An image is received from the image receiving module 110, an image subjected to contraction processing is received from the contraction processing module 232, and logical product processing is performed. Then, the logical product processed image is transferred to the logical sum processing module 241. As a result, an image of only characters other than the ruled lines is created. Further, the character that is in contact with the ruled line is a character in which the contact portion is deleted.

The logical sum synthesis module 140 has a logical sum processing module 241.
The logical sum processing module 241 is connected to the logical sum processing module 227, the logical product processing module 233, and the image output module 150. The logical sum processing image 227 receives an image subjected to logical sum processing (image having only ruled lines) from the logical sum processing module 227, and receives the logical product processing image (image including only characters) from the logical product processing module 233, and performs logical sum processing. Then, the logical sum processed image is transferred to the image output module 150.
FIG. 11 is an explanatory diagram illustrating an example of processing performed by the inversion processing module 231, the contraction processing module 232, the logical product processing module 233, and the logical sum processing module 241. The reversal processing module 231 performs reversal processing on the ruled line image 1000 to create a reversal image 1010, and the contraction processing module 232 performs retraction processing on the reversal image 1010 to create a contact character separation mask 1020. The product processing module 233 performs logical product processing on the received image 300 and the contact character separation mask 1020 to create a character image 1030, and the logical sum processing module 241 performs logical sum processing on the ruled line image 1000 and the character image 1030. To create an output image 1040.

The image output module 150 is connected to the logical sum processing module 241. A dotted line is converted into a solid line from the logical sum processing module 241, and an image separating characters touching the ruled line is received and output.
FIG. 12 is an explanatory diagram illustrating a processing example according to the first exemplary embodiment. A reception image 300 illustrated in FIG. 12A1 is an example of an image received by the image reception module 110. In the region 310, the ruled line and the contact generation regions 311 and 312 are contacted as illustrated in FIG. 12B1. The region 1210 includes a dotted line, a broken line, and an alternate long and short dash line as illustrated in FIG. An output image 1040 illustrated in FIG. 12A2 is an example of an image output by the image output module 150, and the contact portion is separated in the region 1230 where the contact characters exist as illustrated in FIG. 12B2. In the region 1220 including the dotted lines, broken lines, and alternate long and short dash lines (separated regions 1231 and 1232), the solid line is illustrated in FIG.

FIG. 13 is a conceptual module configuration diagram of a configuration example according to the second embodiment. As illustrated in FIG. 13, the image processing apparatus according to the second embodiment includes an image reception module 1310, a dotted line solid line module 1320, a line segment extraction module 1330, a logical product synthesis module 1340, and an image output module 1350. doing.
In the second embodiment, characters are deleted from the received image to form an image with only ruled lines. Unlike the first embodiment, the output image does not have a dotted line as a solid line. For example, it is used when recognizing an image form (original document before characters are described).

The image reception module 1310 is connected to the dotted line solid line module 1320 and the logical product synthesis module 1340. This is equivalent to the image reception module 110 illustrated in FIG. 1, and passes the received image to the dotted line solid line module 1320 and the logical product synthesis module 1340.
The dotted line solid line module 1320 is connected to the image reception module 1310 and the line segment extraction module 1330. The dotted line in the image received by the image receiving module 1310 is made into a solid line. This is equivalent to the processing up to the middle of the table ruled line correction module 120 illustrated in FIG. Details of the configuration, processing contents, and the like in the dotted line solid line module 1320 will be described later with reference to FIG.

The line segment extraction module 1330 is connected to the dotted line solid line module 1320 and the logical product synthesis module 1340. A line segment, which is a ruled line, is extracted from the image that has been solidified by the dotted line solid line module 1320. Then, an image having only ruled lines is passed to the logical product synthesis module 1340.
Specifically, for example, a white run map image and a black run map image are created from an image that has been solid lined by the dotted line solid line module 1320. A line segment is extracted from the image received by the image receiving module 1310 based on the generated white run map image and black run map image. Details of the configuration, processing contents, and the like in the line segment extraction module 1330 will be described later with reference to FIG.

  The logical product synthesis module 1340 is connected to the image reception module 1310, the line segment extraction module 1330, and the image output module 1350. The logical product synthesis of the image of only the ruled line created by the line segment extraction module 1330 and the image accepted by the image receiving module 1310 is performed, and characters are deleted from the document to create an image of only the ruled line. Details of the configuration, processing contents, etc. in the logical product synthesis module 1340 will be described later with reference to FIG.

  The image output module 1350 is connected to the logical product synthesis module 1340. The image output module 150 is equivalent to the image output module 150 illustrated in FIG. 1, and outputs an image received from the logical product synthesis module 1340. For example, the image may be output to the form recognition processing apparatus, and the form recognition processing apparatus may recognize the image and perform processing according to the form type. Alternatively, the data may be output to a table analysis processing apparatus, and the table analysis processing apparatus may perform analysis processing on a table composed of ruled lines.

FIG. 14 is a detailed module configuration diagram of a configuration example according to the second embodiment.
The image reception module 1310 is connected to the white run map image creation module 1421H, the logical sum processing module 1423H, the white run map image creation module 1421V, the logical sum processing module 1423V, and the logical product processing module 1441. This is equivalent to the image reception module 110 illustrated in FIG.

The dotted line solid line module 1320 is divided into a horizontal direction processing module 1400H and a vertical direction processing module 1400V. The horizontal direction processing module 1400H in the dotted line solid line module 1320 includes a white run map image creation module 1421H and a white run map filtering module. 1422H, a logical sum processing module 1423H, and a vertical direction processing module 1400V in the dotted line solid line module 1320 has a white run map image creation module 1421V, a white run map filtering module 1422V, and a logical sum processing module 1423V. ing.
The line segment extraction module 1330 has a logical sum processing module 1434, and other modules are divided into a horizontal direction processing module 1400H and a vertical direction processing module 1400V. The horizontal direction processing module 1400H in the line segment extraction module 1330 Has a white run map image creation module 1431H, a black run map image creation module 1432H, and a black and white run map filtering module 1433H. The vertical direction processing module 1400V in the line segment extraction module 1330 is a white run map image creation module. 1431V, black run map image creation module 1432V, and black and white run map filtering module 1433V.
Both the horizontal direction processing module 1400H and the vertical direction processing module 1400V are different in that they handle horizontal and vertical ruled lines, but the same processing is performed if they are rotated 90 degrees. Have. In the following description, either one or both will be described.

  The white run map image creation module 1421, the white run map filtering module 1422, and the logical sum processing module 1423 in the dotted line solid line module 1320 are respectively a white run map image creation module 222, a white run map filtering module 223 illustrated in FIG. This is equivalent to the logical sum processing module 224.

The white run map image creation module 1431 is connected to the logical sum processing module 1423 and the black and white run map filtering module 1433. A white run map image is created using the image received from the logical sum processing module 1423 (image in which the dotted line is made into a solid line). Then, the created white run map image is passed to the black and white run map filtering module 1433.
The black run map image creation module 1432 is connected to the logical sum processing module 1423 and the black and white run map filtering module 1433. A black run map image is created using the image received from the logical sum processing module 1423 (the image in which the dotted line is made into a solid line). The generated black run map image is passed to the black and white run map filtering module 1433.

The black and white run map filtering module 1433 is connected to the white run map image creation module 1431, the black run map image creation module 1432, and the logical sum processing module 1434. Ruled lines are extracted using the white run map image received from the white run map image creation module 1431 and the black run map image received from the black run map image creation module 1432.
In an image with only a solid line (including an image in which the dotted line is made into a solid line), “The length of the white run length adjacent to the black run length of the character is shorter than the white run length adjacent to the black run length of the solid line. ”. FIG. 15 is an explanatory diagram illustrating a relationship example between a black run length and a white run length. As illustrated in FIG. 15B, the white run lengths of the adjacent regions 1560 and 1570 of the character image 1550 are larger than the white run lengths of the adjacent regions 1520 and 1530 of the linear image 1510 illustrated in FIG. short.

FIG. 16 is an explanatory diagram showing an example of processing by the black and white run map filtering module 1433V. Filtering processing is performed using the white run map image received from the white run map image creation module 1431V and the black run map image received from the black run map image creation module 1432V.
In this filtering process, when the following four conditions are satisfied, the target pixel is output as a black pixel (ON), and in other cases, it is output as a white pixel (OFF).
(1V) The pixel of interest located at the center in the horizontal 9 × 1 pixel filter (filters 1610 and 1620 illustrated in FIGS. 16A and 16B) is a black pixel. (2V) Black run length of the pixel of interest The length is equal to or greater than a predetermined threshold (TH_AV). (3V) The white run length of any one of the four pixels in the right half of the target pixel is equal to or greater than a predetermined threshold (TH_BV). 4V) The white run length of any one of the four pixels in the left half of the target pixel is equal to or greater than a predetermined threshold (TH_CV).

The conditions for processing by the black and white run map filtering module 1433H are also equivalent to the above four conditions. When the following four conditions are satisfied, the target pixel is output as a black pixel (ON), and the others In this case, the pixel is output as a white pixel (OFF).
(1H) The target pixel located in the center of the vertical 1 × 9 pixel filter is a black pixel. (2H) The black run length length of the target pixel is equal to or greater than a predetermined threshold (TH_AH). 3H) The white run length of any of the upper half 4 pixels of the target pixel is equal to or greater than a predetermined threshold (TH_BH). (4H) The white run length of any of the lower half 4 pixels of the target pixel. Is greater than or equal to a predetermined threshold value (TH_CH). Note that whether or not these conditions are satisfied refers to both the white run map image and the black run map image. Further, TH_AV, TH_BV, TH_CV, TH_AH, TH_BH, and TH_CH may be different values or may include the same value.
Note that the size of the filter may be variable so as to correspond to the detected line width.

  The black pixel of the ruled line illustrated in FIG. 16A is output as a black pixel (ON) by the black-and-white run map filtering module 1433V, and the black pixel in the character illustrated in FIG. 16B is output as a white pixel (OFF). Will be output.

The logical sum processing module 1434 is connected to the black and white run map filtering module 1433H, the black and white run map filtering module 1433V, and the logical product processing module 1441. A ruled line in which a dotted line is made a solid line by performing a logical sum synthesis of an image in which only a horizontal ruled line from the black and white run map filtering module 1433H is extracted and an image in which only a vertical ruled line from the black and white run map filtering module 1433V is extracted. Just create an image. Then, the created image is transferred to the logical product processing module 1441.
The logical product processing module 1441 is connected to the image receiving module 1310, the logical sum processing module 1434, and the image output module 1350. An image of only the ruled line including the dotted line is created by performing a logical product process on the image of only the ruled line in which the dotted line from the logical sum processing module 1434 is turned into a solid line and the image from the image receiving module 1310. Then, the created image is transferred to the image output module 1350.

The image output module 1350 is connected to the logical product processing module 1441. An image having only ruled lines is received from the logical product processing module 1441 and output.
FIG. 17 is an explanatory diagram illustrating a processing example according to the second exemplary embodiment. The received image 1710 is an example of an image received by the image receiving module 1310, and includes ruled lines such as characters, photographic images, dotted lines, and solid lines. An output image 1720 is a processing result according to the second embodiment, and is an image output by the image output module 1350. Thus, the image is only a ruled line such as a dotted line or a solid line.

FIG. 18 is a detailed module configuration diagram of a configuration example of the inversion / logical product processing module 1840 according to the third embodiment. This is obtained by replacing the logical product synthesis module 1340 of the second embodiment with an inversion / logical product processing module 1840. In addition, the same code | symbol is attached | subjected to the site | part of the same kind as 2nd Embodiment, and the overlapping description is abbreviate | omitted.
The inversion / logical product processing module 1840 includes an inversion processing module 1841 and a logical product processing module 1842.
The inversion processing module 1841 is connected to the logical sum processing module 1434 and the logical product processing module 1842. An image reversal process is performed only for the ruled lines in which the dotted lines received from the logical sum processing module 1434 are turned into solid lines. The inversion processing is equivalent to the inversion processing module 231 of the first embodiment. Then, the inverted image is passed to the logical product processing module 1842.
The logical product processing module 1842 is connected to the image receiving module 1310, the inversion processing module 1841, and the image output module 1350. By performing an AND process on the image received from the image receiving module 1310 and the inverted image received from the inversion processing module 1841, an image of only the characters from which the ruled lines have been deleted is created. Then, an image of only that character is passed to the image output module 1350.
The image output module 1350 is connected to the logical product processing module 1842. The image received from the logical product processing module 1842 is output. For example, the data may be output to a character recognition device, and the character recognition device may perform character recognition.

  A hardware configuration example of the image processing apparatus according to the above-described embodiment will be described with reference to FIG. The configuration illustrated in FIG. 19 is configured by, for example, a personal computer (PC), and illustrates a hardware configuration example including a data reading unit 1917 such as a scanner and a data output unit 1918 such as a printer.

  A CPU (Central Processing Unit) 1901 includes various modules described in the above-described embodiments, that is, a white run map image creation module 222H, a white run map filtering module 223H, a logical sum processing module 224H, and a black run map image creation module. It is a control unit that executes processing according to a computer program that describes the execution sequence of each module such as 225H, threshold processing module 226H, contraction processing module 232, black and white run map filtering module 1433H.

  A ROM (Read Only Memory) 1902 stores programs, calculation parameters, and the like used by the CPU 1901. A RAM (Random Access Memory) 1903 stores programs used in the execution of the CPU 1901, parameters that change as appropriate during the execution, and the like. These are connected to each other by a host bus 1904 including a CPU bus.

  The host bus 1904 is connected to an external bus 1906 such as a peripheral component interconnect / interface (PCI) bus through a bridge 1905.

  A keyboard 1908 and a pointing device 1909 such as a mouse are input devices operated by an operator. The display 1910 includes a liquid crystal display device or a CRT (Cathode Ray Tube), and displays various types of information as text or image information.

  An HDD (Hard Disk Drive) 1911 includes a hard disk, drives the hard disk, and records or reproduces a program executed by the CPU 1901 and information. The hard disk stores an image received by the image receiving module 110, a run map image, and the like. Further, various computer programs such as various other data processing programs are stored.

  The drive 1912 reads out data or a program recorded in a removable recording medium 1913 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and the data or program is read out as an interface 1907 and an external bus 1906. , A bridge 1905, and a RAM 1903 connected via the host bus 1904. The removable recording medium 1913 can also be used as a data recording area similar to the hard disk.

  The connection port 1914 is a port for connecting the external connection device 1915 and has a connection unit such as USB or IEEE1394. The connection port 1914 is connected to the CPU 1901 and the like via the interface 1907, the external bus 1906, the bridge 1905, the host bus 1904, and the like. A communication unit 1916 is connected to a network and executes data communication processing with the outside. The data reading unit 1917 is a scanner, for example, and executes document reading processing. The data output unit 1918 is, for example, a printer, and executes document data output processing.

  Note that the hardware configuration of the image processing apparatus illustrated in FIG. 19 illustrates one configuration example, and the present embodiment is not limited to the configuration illustrated in FIG. 19, and the modules described in the present embodiment are executed. Any configuration is possible. For example, some modules may be configured with dedicated hardware (for example, Application Specific Integrated Circuit (ASIC), etc.), and some modules are in an external system and connected via a communication line In addition, a plurality of systems shown in FIG. 19 may be connected to each other through communication lines so as to cooperate with each other. Further, it may be incorporated in a copying machine, a fax machine, a scanner, a printer, a multifunction machine (an image processing apparatus having any two or more functions of a scanner, a printer, a copying machine, a fax machine, etc.).

In the above-described embodiment, the processing for the horizontal direction and the vertical direction is shown separately. However, an image rotated by 90 degrees may be processed after performing one processing. Further, the processing for the horizontal direction and the vertical direction may be performed in parallel or in order.
Note that the various embodiments described above may be combined (for example, a module in one embodiment may be applied to another embodiment, replaced, etc.), and the background art may be used as the processing content of each module. You may employ | adopt the technique demonstrated by.
Further, in the description of the above-described embodiment, “more than”, “less than”, “greater than”, and “less than (less than)” in a comparison with a predetermined value contradicts the combination. As long as the above does not occur, “larger”, “smaller (less than)”, “more than”, and “less than” may be used.

The program described above may be provided by being stored in a recording medium, or the program may be provided by communication means. In that case, for example, the above-described program may be regarded as an invention of a “computer-readable recording medium recording the program”.
The “computer-readable recording medium on which a program is recorded” refers to a computer-readable recording medium on which a program is recorded, which is used for program installation, execution, program distribution, and the like.
The recording medium is, for example, a digital versatile disc (DVD), which is a standard established by the DVD Forum, such as “DVD-R, DVD-RW, DVD-RAM,” and DVD + RW. Standard “DVD + R, DVD + RW, etc.”, compact disc (CD), read-only memory (CD-ROM), CD recordable (CD-R), CD rewritable (CD-RW), Blu-ray disc ( Blu-ray Disc (registered trademark), magneto-optical disk (MO), flexible disk (FD), magnetic tape, hard disk, read-only memory (ROM), electrically erasable and rewritable read-only memory (EEPROM), flash Includes memory, random access memory (RAM), etc. .
The program or a part of the program may be recorded on the recording medium for storage or distribution. Also, by communication, for example, a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a wired network used for the Internet, an intranet, an extranet, etc., or wireless communication It may be transmitted using a transmission medium such as a network or a combination of these, or may be carried on a carrier wave.
Furthermore, the program may be a part of another program, or may be recorded on a recording medium together with a separate program. Moreover, it may be divided and recorded on a plurality of recording media. Further, it may be recorded in any manner as long as it can be restored, such as compression or encryption.

DESCRIPTION OF SYMBOLS 110 ... Image reception module 120 ... Table ruled line correction module 130 ... Contact character separation module 140 ... Logical sum synthesis module 150 ... Image output module 221H ... Horizontal direction processing module 221V ... Vertical direction processing module 222 ... White run map image creation module 223 ... White run map filtering module 224 ... logical sum processing module 225 ... black run map image creation module 226 ... threshold processing module 227 ... logical sum processing module 231 ... inversion processing module 232 ... contraction processing module 233 ... logical product processing module 241 ... Logical sum processing module 1310 ... Image reception module 1320 ... Dotted line solid line module 1330 ... Line segment extraction module 1340 ... Logical product synthesis module 1350 ... Image output module 1400H ... Horizontal direction processing module 1400V ... Vertical direction processing module 1421 ... White run map image creation module 1422 ... White run map filtering module 1423 ... Logical sum processing module 1431 ... White run map image creation module 1432 ... Black run map image creation module 1433 ... black and white run map filtering module 1434 ... logical sum processing module 1441 ... logical product processing module 1840 ... inversion / logical product processing module 1841 ... inversion processing module 1842 ... logical product processing module

Claims (4)

Image receiving means for receiving images;
A continuous power image creating means for creating a continuous power image having, as a pixel value, a frequency of continuous black pixels or white pixels in the image received by the image receiving means;
Based on the continuous power image created by the continuous power image creating means, a solid line forming means for performing a solid line to change a line having a break into a solid line; and
The solid line means compares a pixel value of a pixel in the continuous frequency image having a pixel value that is a frequency that is continuous in the horizontal direction or the vertical direction with a predetermined first threshold value. Thus, two first pixels are extracted, and a line segment sandwiched between the two first pixels intersects the direction of the power at a right angle, and is sandwiched between the two first pixels. By comparing the pixel value of the selected pixel with a predetermined second threshold value, the second pixel on the line segment sandwiched between the first pixels is extracted, and the second pixel is An image processing apparatus for performing a solid line as a cut line portion of the cut line . Separating means for separating a pixel block in contact with a solid line or a broken line in the image received by the receiving means based on an image created based on the solid line made by the solid line means The image processing apparatus according to claim 1, further comprising: A white continuous power image creating means for creating a white continuous power image having a frequency of continuous white pixels in the image subjected to solid line processing by the solid line processing means;
A black continuous power image creating means for creating a black continuous power image having as a pixel value a frequency of continuous black pixels in the image subjected to solid line processing by the solid line processing means;
Based on the white continuous power image created by the white continuous power image creating means and the black continuous power image created by the black continuous power image creating means, a line segment is extracted from the image received by the image receiving means. the image processing apparatus according to claim 1 or 2, further comprising a line segment extracting means. Computer
Image receiving means for receiving images;
A continuous power image creating means for creating a continuous power image having, as a pixel value, a frequency of continuous black pixels or white pixels in the image received by the image receiving means;
Based on the continuous power image created by the continuous power image creating means, it functions as a solid line means for performing a solid line to change a broken line to a solid line ,
The solid line means compares a pixel value of a pixel in the continuous frequency image having a pixel value that is a frequency that is continuous in the horizontal direction or the vertical direction with a predetermined first threshold value. Thus, two first pixels are extracted, and a line segment sandwiched between the two first pixels intersects the direction of the power at a right angle, and is sandwiched between the two first pixels. By comparing the pixel value of the selected pixel with a predetermined second threshold value, the second pixel on the line segment sandwiched between the first pixels is extracted, and the second pixel is An image processing program characterized in that a solid line is formed as a cut line portion of the cut line .

Images